构建可模拟企业多任务环境的数字员工系统，解决长期任务协同中的性能退化问题。
CORPGEN: Simulating Corporate Environments with Autonomous Digital Employees in Multi-Horizon Task Environments
- 通过分层规划与分层记忆设计，实现跨任务目标对齐与干扰隔离。
- 在45个以上交织任务中完成率提升至15.2%，是基线4.3%的3.5倍。
- 适合研究长期决策、多任务协同或企业自动化系统的开发者参考。
长周期推理是自主智能体的核心挑战，但现有基准仅评估单一任务孤立表现。真实组织工作需同时处理多个长周期任务，涉及交错执行、依赖关系和动态重优先级。本文提出多时域任务环境（MHTEs）：一种要求在持续运行环境中协调数十个交错任务（45+，每任务500-1500+步）的新问题类别。我们识别出四种导致基线智能体完成率从16.7%降至8.7%的失效模式（负载从25%增至100%），包括上下文饱和（复杂度由O(1)升至O(N)）、记忆干扰、依赖复杂性（有向无环图对比链式结构）及重优先级开销。为此提出CorpGen框架——一种架构无关的解决方案，包含分层规划、子代理隔离、分层记忆（工作、结构化、语义）与自适应摘要机制。CorpGen通过具有持久身份与真实日程的数字员工模拟企业环境。在三个CUA后端（UFO2、OpenAI CUA、分层）与OSWorld Office数据集上，相比基线（4.3%）实现最高15.2%的完成率，提升达3.5倍，且性能稳定随负载增长。消融实验表明经验学习贡献最大。
原文摘要 · Abstract (English)
Long-horizon reasoning is a key challenge for autonomous agents, yet existing benchmarks evaluate agents on single tasks in isolation. Real organizational work requires managing many concurrent long-horizon tasks with interleaving, dependencies, and reprioritization. We introduce Multi-Horizon Task Environments (MHTEs): a distinct problem class requiring coherent execution across dozens of interleaved tasks (45+, 500-1500+ steps) within persistent execution contexts spanning hours. We identify four failure modes that cause baseline CUAs to degrade from 16.7% to 8.7% completion as load scales 25% to 100%, a pattern consistent across three independent implementations. These failure modes are context saturation (O(N) vs O(1) growth), memory interference, dependency complexity (DAGs vs. chains), and reprioritization overhead. We present CorpGen, an architecture-agnostic framework addressing these failures via hierarchical planning for multi-horizon goal alignment, sub-agent isolation preventing cross-task contamination, tiered memory (working, structured, semantic), and adaptive summarization. CorpGen simulates corporate environments through digital employees with persistent identities and realistic schedules. Across three CUA backends (UFO2, OpenAI CUA, hierarchical) on OSWorld Office, CorpGen achieves up to 3.5x improvement over baselines (15.2% vs 4.3%) with stable performance under increasing load, confirming that gains stem from architectural mechanisms rather than specific CUA implementations. Ablation studies show experiential learning provides the largest gains.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。